[PATCH] [XFRM/IPV6] fix next header offset in decode_session

[PATCH] [XFRM/IPV6] fix next header offset in decode_session

[Jean-Philippe Andriot]

The way to get the next protocol number of an IPv6 tunnel changes
after introducing IP6CB, but I think we should go back to the
previous version here.
In our case I think there was a confusion between the pointer
on the first byte of the next header and the value of the next
header field.

Signed-off-by: Andriot Jean-Philippe <jean-philippe.andriot@6wind.com>

--- xfrm6_policy.c.org  2006-11-07 09:45:47.000000000 +0100
+++ xfrm6_policy.c      2006-11-07 09:46:19.000000000 +0100
@@ -255,7 +255,7 @@ _decode_session6(struct sk_buff *skb, st
        u16 offset = skb->h.raw - skb->nh.raw;
        struct ipv6hdr *hdr = skb->nh.ipv6h;
        struct ipv6_opt_hdr *exthdr;
-       u8 nexthdr = skb->nh.raw[IP6CB(skb)->nhoff];
+       u8 nexthdr = skb->nh.ipv6h->nexthdr;
 
        memset(fl, 0, sizeof(struct flowi));
        ipv6_addr_copy(&fl->fl6_dst, &hdr->daddr);

Re: [PATCH] [XFRM/IPV6] fix next header offset in decode_session

[YOSHIFUJI Hideaki / 吉藤英明]

In article <20061107093002.GA6576@palpatine.dev.6wind.com> (at Tue, 7 Nov 2006 10:30:02 +0100), Jean-Philippe Andriot <jean-philippe.andriot@6WIND.com> says:

> The way to get the next protocol number of an IPv6 tunnel changes
> after introducing IP6CB, but I think we should go back to the
> previous version here.
:
>         struct ipv6_opt_hdr *exthdr;
> -       u8 nexthdr = skb->nh.raw[IP6CB(skb)->nhoff];
> +       u8 nexthdr = skb->nh.ipv6h->nexthdr;
>  
>         memset(fl, 0, sizeof(struct flowi));

I disagree.

If you do this, you refer to the first extension headers only.
We need to skip preceding extension headers using IP6CB(skb)->nhoff,
which holds the offset to the current nexthdr.

--yoshfuji